JD Speeds Up Image Analysis by Replacing GPUs with CPUs

JD is a leading retailer and cloud service provider in China. To analyze its huge database of merchandise images, JD worked closely with Intel to deploy the BigDL deep learning library on the Intel® Xeon® processor-based servers that store the images. It also used The Apache Hadoop* and Spark* frameworks handle resource management, making it easier to develop new applications in the future, while still remaining performance efficient

After making these upgrades, JD found that

  • Performance increased by up to 3.83×1, compared to the alternative GPU based solutions
  • JD established a platform for innovation that will be used to create new applications easily, for internal use and new public cloud services
  • JD lowered its total cost of ownership for the solution, by reusing its existing hardware estate for analysis, compared to operating a separate GPU cluster